INDEX

Explanations

mathematical symbols and expressions related to derivatives and integrals

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sse

-0.16

æľŃ

-0.15

 Relay

-0.15

 Pert

-0.14

eten

-0.14

 figure

-0.14

relay

-0.14

Î´Î¯

-0.14

Ã¡Å¡

-0.14

áº¿t

-0.14

POSITIVE LOGITS

ĶåĽŀ

0.16

readcr

0.15

lyn

0.15

oproject

0.14

witch

0.14

à¸ķà¸£

0.14

lettes

0.14

 Boxing

0.14

0.13

idon

0.13

Activations Density 0.032%

mathematical symbols and expressions related to derivatives and integrals

No Comments

No Known Activations